More than three decades ago, Albert Hirschman wondered why not enough attention had been paid to Latin America’s post-war economic boom, certainly not while it happened.Footnote 1 Only in retrospect, from the “lost decade” of the 1980s, did the accomplishments of the developmental era become relatively clear. But even then, the comparison with East Asia that began during the 1980s diminished the region’s achievements.Footnote 2 If anything, the period from 1945 to 1975 came to be seen as a lost opportunity, when Latin America supposedly took the wrong direction while East Asia cleared the path to its economic success. This perspective could not be more different than the attention paid to the “Spanish Miracle” – with its own Wikipedia entry! Spain’s developmental achievements were seen as a victory for the Franco regime, whose replacement by a vigorous democracy made those achievements even better.
Much of the criticism of the Latin American cases was directed at the supposed failure of the developmental state in the continent. Developmental policies came to be seen as obstacles to the economic progress of the region, leading finally to the debt crisis of the early 1980s, and the subsequent collapse of the Latin American economic model.
In the present volume, we will examine both the achievements and failures of developmental states in Latin America and Spain. We will seek to clarify how developmental state institutions were conceived in the context of far-reaching visions for national affirmation and progress, and how they were designed and established since the early 1930s. We will analyze what the developmental states tried to do, and why they succeeded or failed at their many diverse tasks. We intend to use the contrasting views of the national cases under discussion as a source to continue our long-term investigation, begun with Republics of the Possible.Footnote 3 That previous volume discussed the politics and techniques of state and nation making in Latin America and Spain during the long “liberal era” from 1810 to 1930. The first volume on the liberal state, and the second or present volume on the developmental state, can be read separately without any loss of content or perspective, since they represent complete units by themselves. Nevertheless, the editors and the authors have followed some key issues, and employed some common categories of research, which we will briefly describe further below. At the conclusion of the first volume, we realized that while we could find significant variances between the countries under examination, a similar narrative could be told of the creation of a liberal order founded on domestic privilege and international dependence across the Hispanic-American world.
The Great Depression shattered that status quo. In the case of Spain, the subsequent Civil War, 1936–1939, left an economy much smaller and less vibrant than before, and a devastated society. The Spanish state after the Civil War, while not particularly strong, did enjoy a monopoly over political power, and it played a central role in defining economic policy. While Franco remained head of state until 1975, Francoism can be divided into two periods. During the first two decades of the regime, the government was dominated by Falangist and corporatist military officers and politicians, who favored an autarchic economy. Economic difficulties and the need to reach out to the rest of the world required a shift in power to a new generation of technocrats, especially after 1959. It was only in the 1960s that Spain became “different” from its own Black Legend, and began to acquire a new reputation as a dynamic society.
While Latin America largely avoided military conflicts – with the most prominent exception of the Chaco War, 1932–1935 – the collapse of commodity markets in the 1930s led to broad economic decline, and produced new social and political tensions. The post-depression era, however, also witnessed a growth in state capacity and the completion of the process of institutionalization initiated during the nineteenth century. Interestingly, Latin America began its return to prosperity much earlier than Spain. It was only in the late 1950s that the Spanish economy returned to 1929 levels, after spending almost three decades poorer than most of the continent across the Atlantic. It was not until the 1960s that Spain’s national income per capita began to outpace the major Latin American economies.
Even during that decade, the cultural dynamism of Latin America could contrast with the conservative repression of Franco’s Spain. The 1960s represented the zenith of the Latin American (almost uniformly white) middle class. Internationally, the continent was in the midst of economic expansion. Just as importantly, the region was a cultural and intellectual powerhouse as represented by the literary “boom” and the high regard in which Latin American artists, thinkers, and lifestyles came to be held. Yet, the global centrality of Latin America was temporary. Consider that in 1968, when Mexico hosted the summer Olympics, its economy was roughly the same size as Spain’s. Twenty years later, the Olympics were held in Seoul, and Mexico was poorer than both Korea and Spain. By 1992, when the Olympics were held in Spain, the gap had grown even larger. By 2014, when the Olympics returned to Latin America, Spain and Korea’s per capita income was more than double that of Mexico and Brazil. While the latter two’s economies had largely flatlined, Spain and Korea had grown exponentially.
Why? There is not a student of Latin America who on visiting Madrid or Seoul does not ask themself the same question. The too-often heard explanation of “Confucian” discipline as opposed to “Latin” self-indulgence does not survive the inclusion of Spain in the comparison. What was it about the developmental state in Latin America that slowed growth, beginning in the late 1960s?
The Developmental State Concept
Before we address in detail the issue of the consistent deceleration of economic growth since the late 1960s in Latin America, and its marked contrast to increasing growth in Spain, we will introduce the general concept of a developmental state. We will also describe, in this section, the dimensions or categories of state capacity that we employ as analytical tools in the book.
According to Woo-Cumings, the developmental state is a “seamless web of political, bureaucratic, and moneyed interests that structures economic life.”Footnote 4 The term was originally used to describe the Asian “Tigers” and primarily Japan, followed by Korea and Taiwan. Intended as an argument against those who attributed Northeast Asian success to following market-centered economic policies, the concept involves a national commitment to development guided by career bureaucrats controlling state finances allied with private industry. The Latin American equivalent can be seen as sharing several characteristics: investment decisions made by the state, a “developmental discourse,” and the partial exclusion of the popular sector (depending on regime type). Importantly, and as we will see, significantly, it does not include in a consistent way the Weberian, professional bureaucracy common to Northeast Asian countries.
For the study of the developmental state, we employ in the present volume four categories of research, which we define as dimensions of state capacity or strength: infrastructural, territorial, economic, and symbolic. The different parts of the book, as shown in the table of contents, are organized in relation to their focus on one or more of those categories. We employed the same pattern of analysis, based on four dimensions of state strength, for the first book on state and nation making in Latin America and Spain, published by the same editors a few years ago.Footnote 5 As mentioned above, the earlier volume discussed the politics and techniques of state building during the long liberal era from 1810 to 1930. The two volumes represent complete units by themselves, but the editors and authors have followed some central issues, and applied conceptual categories that connect both projects.
Sociology and political science frequently employ the concept of state capacity and related terminology and ideas, such as strength, power, and influence. However, the notion of state capacity became a regular part of developmental literature only in the 1980s. Ideas such as strength and power are deceptively simple: the problem comes from attempting to use them in a systematic manner across a variety of cases. What is it that states do, and how can we trace the transformation of their various capacities across time in Latin America and Spain? Combining a variety of typologies, from Weber to Bourdieu and Mann, we propose four different categories or dimensions of state strength.
The first category of state capacity that we employ is based on the notion of infrastructural power, as originally introduced by Michael Mann.Footnote 6 According to Mann, infrastructural power refers to the capacity of the state to coordinate society by means of the diffusion of law and administration in many areas of social life, which had remained outside the scope of state concern before the vast expansion of this state capacity during the second half of the nineteenth century. Infrastructural power involves organizational and technical skills to collect and process information, build organizational structures, and maintain communication and interaction networks. Infrastructural power is a key dimension, because this is what makes modern states exceptionally strong.Footnote 7
The expansion and diversification of bureaucratic organizations increases the penetration of the state in terms of infrastructural power. However, according to Mann, such increase of infrastructural power does not imply, as Weber mistakenly assumed, more vertical concentration of power in a central authority. Infrastructural capacity does not involve centralization of power; rather, the contrary is the case. First of all, modern state administration “almost never forms a single, bureaucratic whole.”Footnote 8 The infrastructure of the modern state is formed by an array of bureaucratic organizations variously linked to power networks in civil society. Secondly, the expansion of infrastructural penetration predictably goes both ways: as a result of the embeddedness of relatively autonomous bureaucratic organizations, civil society’s capacity to bring influence to bear on the state also increases. The expansion of infrastructural power occurs simultaneously with the widespread politicization of civil society.Footnote 9
The second dimension of state strength we call territorial power, and it involves the classic Weberian notion of monopoly over the means of violence. Note that we explicitly do not specify the legitimate use of that violence as we wish to distinguish between a simple capacity to coerce from the much more complex notion of justifying such coercion. Mann also called this category of power despotic, and it represents the influence that state elites are able to exert over the population of a certain area, without having to enter into routine negotiations with other actors. The concept of despotic power captures the conventional perception of power as the capacity to issue and impose commands. This form of state power or capacity is the simplest to wield, as it merely requires the acquisition and utilization of enough relative coercive force to impose order on a certain territory. This is the state as disciplinary institution, and it takes place on two fronts: first, in relation to other states defining sovereignty; and secondly, against internal or domestic rival claimants and subjugated groups.
Before we consider the next two dimensions of state strength, economic, and symbolic, we will briefly examine the close interconnection of the first two categories during the developmental era. In the context of national development strategies, infrastructural and territorial power were thoroughly articulated by long-term institutional projects. In contrast, during the previous period of state building, the liberal era both in Latin America and Spain, territorial power in itself was a predominant concern of state elites, and they attempted to consolidate this capacity by creating and deploying military and police forces, including custom guards, in areas close to national borders and in rural spaces – the prominent and controversial Spanish Guardia Civil was created for this purpose in 1844. However, during the developmental era, the states focused instead on increasing their dominion over peripheral territories by creating new bureaucratic organizations, which took the form, on the one hand, of many regional developmental agencies created in Latin America since the 1940s. As unitary states, on the other hand, Chile and Spain avoided the creation of regional developmental agencies. However, the promotion of economic development by central agencies, in both countries, was deliberately targeted on peripheral regions, as shown by the case of Catalonia, for example, where substantial projects of state-led industrialization under Francoism were located.
Developmental strategies included from the beginning plans to achieve the modernization and professionalization of public bureaucracies, that is to say, plans to increase the overall infrastructural strength of the state. Almost in every case, moreover, the professionalization of the central civil service was carried out together with the creation of semi-autonomous developmental agencies. Those semi-autonomous developmental agencies were conceived and designed as institutions of superior bureaucratic quality, compared to the rest of the public administration. The concept of “islands of development” was created by Thurber to define precisely those “nuclei of strength, especially organizational strength” that developmental agencies represented.Footnote 10 In sum, the many regional development agencies created in Latin America, or the central autonomous development agencies created by Chile and Spain, were designed with the goal of increasing infrastructural power or bureaucratic capacity and, at the same time, improving the territorial – regional – reach of the state, its territorial strength. Since infrastructural and territorial power were so closely intertwined during the developmental era, the corresponding part of the book – chapters five to eight – examine those two dimensions of state strength in combination.
We define the third category of state capacity as economic power, and this involves diverse connected processes. First, economic power is about the state promoting the general prosperity of a society. Prior to the Keynesian revolution, states mostly contributed to prosperity in the course of the unification of an economic space through the creation of a national market. Of greater relevance for our cases, the states may also increase prosperity by creating the physical and legal infrastructure supporting the insertion of their domestic economy into a global system of exchange. A second aspect of economic power involves the control over and appropriation of resources through the establishment of an efficient tax system. The third and perhaps most extensive aspect of economic power, during the developmental era, concerns the formulation and implementation of long-term economic policies, particularly industrial promotion, welfare and labour services and regulations, public credit, trade strategies, and others.
The fourth dimension of state capacity is what Bourdieu calls symbolic power or what Weber discussed as legitimacy. As Bourdieu notes, “what appears to us today as self-evident, as beneath consciousness and choice, has quite often been the stake of struggles and instituted only as the result of dogged confrontations.”Footnote 11 The study of the state’s symbolic power is the history of how it attempts to construct its own sense of inevitability. Symbolic strength is the quality that should – ideally – place the authority of the state out of the bounds of contention. Regarding this symbolic dimension, Joseph Strayer assigns a central role to what he calls “loyalty” during the consolidation of state power, a “shift in the scale of loyalties” from earlier societies, and a new “priority of obligation” towards public institutions, or what he later calls a “cult of the state.”Footnote 12
During the developmental era, new and powerful narratives of the national community emerged, and redefinitions of citizens’ political and ethnic identities were attempted in several national cases. States were eager, of course, to take control of national narratives, and to position public institutions and official practices as stages for the performance of citizens’ political identities, in order to expand and strengthen mass loyalties. Developmental projects were heavily invested with the symbolism of national destiny, while at the same time they were often conceived and carried out by new social and political actors that demanded to enter the public scene. States and public agencies tried to appropriate the symbolic potential emerging from those social and cultural transformations, by their own self-presentation as agents of change, with more or less success in the diverse national cases.
Developmental Success Compared
Why did some developmental states succeed, occasionally beyond all expectations, while others did not? The most popular explanations examine Latin America and Northeast Asia, arguing for instance that the geopolitical situation in Asia provided significant political will, as well as ample funding, resulting from the West’s grand strategy during the Cold War. One aspect that has also drawn attention is the extent of policy continuity. For many, the consistency of the Northeast Asian model is critical. Others, however, point out that Latin America adhered for too long to the model of import-substitution industrialization (ISI). There was also a long debate about the extent to which democracies could not solve the collective action problems of rapid development, but this relationship was found to be generally weak.Footnote 13 Another interesting approach comes from Fajnzylber’s critique of the “showcase modernity” in Latin America, which he argues was too focused on providing a middle-class lifestyle to urban professionals, as opposed to more effective policies of capital accumulation.Footnote 14
The case of Spain provides a significant empirical comparison in this whole discussion. As mentioned above, the Spanish economy lagged behind several major Latin American countries at the middle of the twentieth century. The association of the developmental state with authoritarian policies was a feature of both the Latin American and Spanish cases during the 1960s, and the divergent fortunes of Spain and its former colonies were actually less predetermined than intuition suggests. In 1960, it was not obvious which country of the Iberian world would be better off by 1990. In the pages that follow, we will begin to address this question by comparing the Latin American and Spanish cases across diverse empirical measures.
Economies and Societies
We begin with a well-known tale of economic history. In 1995, Spain’s real gross domestic product (GDP) per capita was $12,860, almost $5,000 more than Argentina’s and over double that of Mexico and Brazil, an apparently clear reflection of the global North–South divide.Footnote 15 But at the middle of the twentieth century, Spain’s economic performance compared to that of Latin America seemed mediocre at best. Even before the Civil War ravaged its economy, Spain was far behind Argentina in terms of real GDP per capita. In 1925, Argentina’s output was nearly $4,000 per capita (in 1990 dollars), compared to under $2,500 for Spain. Latin American economies would develop more rapidly for years to come. Between 1925 and 1960, real output per capita grew by 40 percent in Argentina, 65 percent in Mexico, and 133 percent in Brazil. Spain’s output, meanwhile, increased just 25 percent during the same period.
It was only afterwards that the familiar North–South divide took shape. Figure 1.1 illustrates how, after decades of mediocre growth, Spain’s real GDP per capita accelerated at an average rate of 9.1 percent every year between 1960 and 1995. By contrast, the Latin American economies – particularly Argentina’s – floundered, in many cases actually losing ground.
Figure 1.1 Real GDP per Capita, 1925–1995, for Spain and Select Latin American Countries, in 1990 Geary-Khamis (GK) dollarsFootnote 16
The solid black line representing Spain sticks out on the chart. The drastic reversal of fortunes holds even when comparing Spain and Mexico to their highly developed neighbors. Despite high economic growth in Mexico between 1925 and 1960, the economy of the United States consistently remained twenty times as large. Events in the second half of the century, including industrialization, and the 1994 signing of the North American Free Trade Agreement (NAFTA), had a relatively modest impact: by the 2000s, the Mexican economy had settled at 9 percent the size of its American counterpart. As for Spain, when the Civil War ended in 1939, its economy was less than half the size of the Western European average, a ratio that only modestly improved by 1960. If Spain’s growth could be explained by the post-war European boom, then we would expect this ratio to remain fairly stable. Instead, when the European Union formed in 1988, Spain’s economy reached parity with the continental average; by the early 2000s, it surpassed the Western European mean by a third.
Clearly, some profound transformation occurred in Spain during the 1960s. The economic boom has been associated with the authoritarian government’s decision to replace, in leading public policy positions, old Falangists with young, highly influential technocrats.Footnote 18 Afterwards, the Spanish developmental state was very successful at promoting industrial production and export growth. Why similar Latin American attempts ended in disaster is less obvious. As we have seen, not only did the continent enjoy high levels of economic development in the first half of the twentieth century, but Latin American rulers emulated Spain’s developmental model without delay. As early as the mid 1960s, the military dictatorships of Argentina and Brazil were eager to hire expert managers from Opus Dei, the same right-wing Catholic organization that trained Spanish technocrats.Footnote 19 And yet when these nations eventually emerged from military rule, only Spain could claim to have witnessed substantial developmental achievements.
Figure 1.2 plots Spain’s GDP per capita as a ratio of the Latin American mean. Average Spanish citizens were earning just 60 cents to the dollar of their counterparts across seven Latin American countries (LA7: Argentina, Brazil, Chile, Colombia, Mexico, Peru, and Venezuela) between 1935 and 1960. One generation later, they were making double. In other words, the Spanish developmental success is made more remarkable by the conspicuous absence of any Latin American equivalent.
Later chapters and the conclusion of this volume examine in detail why the developmental state did not reach any extraordinary results in Latin America, as it did in Spain. Before discussing the issue, however, let us define here the heart of this comparison: the inflection point circa 1960, when the Spanish and Latin American growth trajectories diverged. In the rest of this section we compare historic trends in Spain and Latin America across further empirical measures associated with developmental results. The broad period of analysis, from 1925 to 1995, provides thirty-five years of observations on either side of the inflection point. We examine economic, demographic, social, and political trends for peculiar historical circumstances that may help explain Spain’s very fast economic growth. Our aim is to examine common explanations of developmental success (or failure) while understanding the inflection point in more detail. Two patterns stand out in particular: one, that for most of the twentieth century, Spain tracks closely with the Latin American average across a wide range of developmental measures; and two, the inflection point appears in the data with surprising frequency, albeit at different periods across measures.
Manufacturing and Trade
One plausible explanation for Spain’s fast industrial growth during the 1960s is that Spain was simply more industrialized to begin with. The data, however, suggest otherwise. As late as 1970, Spain lagged behind Argentina, Chile, and Venezuela in its share of GDP per capita originating in industrial activity, a measure on which it historically resembled the Latin American average.Footnote 20 Over the next decade, however, Spain experienced a near-exponential rate of industrialization that dwarfed the linear growth paths of Latin American economies. By 1981, Spain’s industrial GDP per capita had risen eightfold to more than double the level in Argentina, and two and a half times as large as the Latin American mean.
Crucially, Latin American nations did not “fail” to industrialize – they simply did so at linear rates that were no match for Spain’s exponential pace. An identical trend occurs in the per capita production of steel and cement, a historic industrial activity easy to compare over time and space.Footnote 21 Spanish production levels were essentially flat between 1925 and 1955, lagging behind Chile and Venezuela. They began skyrocketing soon afterward, reaching a peak in the 1980s, when Spain annually produced 2.2 megatons of steel and cement per capita, thrice as much as the largest producers in Latin America, Brazil, and Mexico. Although these states significantly increased their output during this period, it remained a fraction of Spain’s.
If Spain wasn’t historically more industrialized, was it at least more integrated into world markets? After all, Spain was an imperial power that conducted vast military and economic operations for centuries. Perhaps trade linkages helped to “unlock” the growth potential of the Spanish economy. In fact, between 1925 and 1962, the average annual share of world trade among Argentina, Brazil, and Mexico (LA3) was 56 percent larger than Spain’s.Footnote 22 This difference was especially pronounced during wartime: in 1937, Spanish trade was worth just 14 percent of Argentina’s share and less than half of Mexico’s (and it stayed that way until the 1950s). Here the data reveal another inflection point: Spanish international commerce surpassed LA3 in 1963, accounting for an even larger share of global trade throughout the rest of the twentieth century.
Spain’s trade boom occurred in the early years of the Spanish period of intensive growth, suggesting that trade policy (along with steel and cement production) was central to the development strategies implemented by the technocrats. For example, absolute levels of Spanish exports often lagged behind two major commodity-rich Latin American economies – Brazil and Mexico – until at least the late 1980s. Behind the scenes, however, Spain transformed from a commodity-exports to a manufacturing-exports regime. The manufacturing share of Spanish exports (as a percentage of all exports) grew from just under 30 percent in 1962 to 73 percent in 1988. By contrast, value-added manufactured goods accounted for only 57 percent of all 1988 exports from Brazil and 49 percent of exports from Mexico. The changing nature of Spanish exports is thus an important qualitative distinction that can be directly linked to developmental strategies.
One critical aspect of trade provided Spain with a key source of foreign currency. The Net Travel and Tourism Balance (NTTB) calculates a country’s revenues from travel and tourism after deduction of the money its nationals spend on travel abroad.Footnote 23 In 1995, Spain’s NTTB was $20.8 billion, ten times greater than Mexico. But in 1960, Mexico’s travel balance, at $260 million, was actually higher than Spain’s. While tourism to Mexico suffered during the peso crisis of the 1980s, it expanded eighty-fold in Spain. Like other forms of trade, Spain’s travel balance did not begin at high levels but rather hit an inflection point in the early 1960s.
Development for Whom?
If Spain had no clear initial advantage over Latin America in growth, industrialization, or trade prior to 1960, then we might want to narrow our scope and look at aspects of human development. For instance, developmental states (especially in East Asia) are associated with relatively low inequality. Did Spain start off more highly equal than Latin America, possibly impacting on the relative success of developmental strategies? The twentieth-century tale of inequality in Latin America is a turbulent one with no clear narrative – for most of the second half of the century, the Gini coefficients of the major Latin American countries oscillated wildly between 40 and 60, although the continent as a whole ended the century with greater income inequality than it had mid century.Footnote 24 Surely Spain was more egalitarian? In fact, available data shows that Spain was slightly more unequal than Latin America until the 1960s! By the 1970s, however, Spain’s Gini coefficient dropped dramatically relative to the LA7 median (a more sensible measure than the mean due to nation-wide differences in Gini calculation). In 1973, for instance, median income inequality in Latin America was 30 percent higher than in Spain. And by 1980, Spanish inequality dropped to half of the LA7 median (although the gap later narrowed). Surprisingly, this relative performance was driven largely by rising equality in Spain rather than increasing inequality in Latin America.
We can still imagine that Spain enjoyed an early comparative advantage in other essential developmental factors, like education. Perhaps Spain started off with a more highly educated population than Latin America, resulting in a high stock of human capital, which enabled the Spanish developmental state to reach the extraordinary economic success that remained missing in the Latin American cases. Yet again, the data reveal a post-1960s inflection point: relative to Latin America, education levels in Spain grew only after this time. In 1970, Argentina, Peru, and Chile each had significantly more university students per 10,000 population than Spain.Footnote 25 Through most of the 1960s, Argentina had a higher rate of tertiary education than Spain, Brazil, and Mexico combined. And while university attendance in Spain grew rapidly after 1970, increasing sevenfold by century’s end, it expanded at impressive rates across Latin America as well. Tertiary education rates in Argentina and Peru, particularly, tracked closely with Spain until the 1990s. We can be fairly confident that the shortcomings of the developmental state in Latin America were not caused by lack of higher education.
Spain had no initial advantage over Latin America in secondary schooling either, but these trends provide more nuanced evidence: we find here differences in general developmental strategies. By 1965, when its university attendance still lagged far behind Argentina’s and Chile’s, Spain became the leader in secondary school attendance, and continued to grow at a rate outpacing any country in Latin America, where enrollment tended to grow in quick bursts before suddenly leveling off. As of 1990, secondary school attendance in Spain was one and a half times more than that of Mexico and Peru, nearly twice the levels of Argentina and Chile, and nearly five times as much as Brazil – a puzzling development considering that university attendance in Argentina and Peru more or less equaled Spain that year. This suggests that while Spain emphasized democratizing education early on, it remained largely a middle-class domain in Latin America. Expanding access to secondary schooling had important consequences for the creation of Spain’s skilled workforce. But as we shall see in a later section, while this may have been a concerted effort on behalf of Spanish authorities, they also benefited from demographic advantages beyond the reach of any Latin American state.
To be fair, Spain certainly had some early-stage developmental advantages over Latin America, particularly in two important arenas: infant mortality and access to health care. Spain lagged only behind Argentina on both measures in the 1920s. Moreover, Spain’s progress on infant mortality preceded the Spanish period of fast economic growth, for once. During the war, infant mortality rates in Spain spiked to higher levels than most LA7 countries, but recovered quickly under the old Falangist guard (a period, we might recall, of otherwise sluggish development). By the mid 1940s, infant mortality in Spain dropped from about 145 to roughly 60 deaths per 1,000 live births – lower even than Argentina.Footnote 26 Latin American states made commendable progress on infant mortality in their own right. By 1995, Chile’s infant mortality rate dropped to 12 deaths per 1,000 live births from a high of over 250 earlier in the century; Mexico’s rate dropped to 16 from a height of over 200. Nevertheless, Spain’s rate by then dropped to below 6 deaths per 1,000. Ironically, the least progress on mortality was made by the former leader, Argentina, which by 1995 ranked next to last (after Peru), with a rate of 23 deaths per 1,000.
Argentina’s relatively poor performance was not equivalent to lack of access to health care. The number of physicians per capita is frequently used by the World Health Organization (WHO) and other organizations as a measure of health care availability.Footnote 27 In the mid 1940s, both Spain and Argentina had one doctor per 1,000 residents, nearly twice as many as Chile and Mexico. While Latin America’s access to health care improved slowly and steadily between 1946 (the first year this measure is available for most countries) and 1995, Spain’s ratio virtually flatlined at one doctor per 1,000 between 1946 and 1960. The reader will not be surprised to learn that the logic of the inflection point also applies to doctors in Spain, whose numbers rose dramatically shortly after 1960. By 1994, Spain had four doctors per 1,000 residents – thrice as many as the Latin American average, four times the level of Mexico, and much higher than Argentina’s rate of 2.63.
Middle-Class Creation
In sum, overall comparison with Spain shows that Latin American states were by no means developmental failures. Rather, it demonstrates steady progress on important measures of manufacturing, trade and human development throughout most of the twentieth century. On some measures, like higher education, Latin American states made very significant progress on the whole. What the inflection point tells us, though, is that Spain somehow pivoted from mediocre to exponential progress on all of these measures after 1960 (with the notable exception of infant mortality).
From our vantage point, the period of fast economic growth in Spain looks like a developmentally self-reinforcing cycle, with substantial improvements in health, human capital stock, and industrial infrastructure, which all served to supercharge each other, and the economy as a whole. But Spain was a dark horse lacking any obvious comparative advantage. If anything, Argentina was the clear favorite in the developmental race, maintaining a wide lead on most measures for most of the twentieth century. But when the Argentine bureaucratic-authoritarian state tried to copy Spain’s technocratic model, it faltered. So did other Latin American states that closely resembled Spain prior to the 1960s.
What happened? Now may be a good time to revisit Fajnzylber’s “show-case modernity” critique of Latin America. Did the continent squander its potential on middle-class urban professionals? If that were indeed the case, then we would expect not to see the Spanish inflection point in measures of middle-class lifestyle. We begin by analyzing the consumption patterns of air traffic, phones, and automobiles. We selected these as they can be used as comparative levels of middle-class consumption. Yet here again we see the same patterns as before. From 1929 to 1959, Spain and LA7 had roughly comparable levels of commercial air traffic with the exception of Mexico, where air travel was much more common.Footnote 28 But by the early 1960s, commercial air travel began to grow exponentially in Spain. By the late 1960s, air traffic in Spain pulled ahead of Mexico, and far ahead of the LA7 average, which proceeded to rise at a more or less linear rate. Similarly, in the 1920s, Spain had fewer phones per capita than the Latin American average – 10.5 connections per 1,000 population for Spain in 1927, compared to LA7’s mean of 11.92.Footnote 29 But by 1967, Spain had well over three times as much landline penetration, and more than 4.5 times as many per capita landlines by 1977. Between 1948 and 1984 phone penetration in Spain grew by 2,027 percent – averaging 50 percent per year.
Given that these empirical trends closely follow cross-national patterns in more “productive” measures, perhaps we should re-examine the “showcase modernity” conclusion. A closer look at the case of automobiles, however, does lend credence to Fajnzylber and the import-substitution hypothesis. Until 1960, Argentina, Mexico, and Brazil together boasted higher rates of car ownership than Spain. There were 2.8 cars for every 1,000 people in Spain in 1960, and 5 per 1,000 on average in LA3 (2.85 for LA7).Footnote 30 But by the mid 1990s, Spain and Brazil had nearly twice as many cars per capita as Mexico, and 3–4 times as many as Argentina. If in Mexico and Brazil the Volkswagen Beetle was the automotive symbol of development, the home-grown SEAT 600 was the vehicle of Spain. Most importantly, while most Latin American auto industries (with the exception of Mexico) produced largely for the domestic market, and all were owned by multinationals, Spain’s auto industry included a significant share of national capital and public ownership, and it was able to expand substantially into export markets.
Demography
We haven’t yet mentioned a critical structure component that sets Spain apart from Latin America independently of any inflection points: people. Population growth in Latin America far outstripped Spain in the twentieth century. Even the LA7 country with the least rate of population increase, Argentina, grew nearly three times as much as Spain between 1925 and 1995.Footnote 31 Not only was Brazil always the most populous country among LA7 and Spain, but it also experienced a high population growth rate relative to all other countries under comparison, growing especially rapidly post-1960. This is also true of Venezuela, Colombia, and Mexico, which unseated Spain as the second-most populous country in 1954.
Where the population concentrated was also markedly different. Among LA7 and Spain, Argentina and Chile had the highest percentage living in their major city. Over 30 percent of the Argentinian population lived in Buenos Aires as early as 1960, a trend that continues until at least 1990; in Chile, nearly 38 percent of the population had moved to Santiago by 1970, an aggressive rate of capitalization that crossed 40 percent by 1990. Peru too saw a rapid rate of capitalization, with Lima accounting for 3.3 percent of the national population in 1920, and 30.2 percent by 1990. On the other hand, only 3.5 percent of the Spanish population lived in Madrid in 1920, and this figure rose modestly to 7.6 percent by 1990. Migration is another interesting comparison. The big picture is that Spain made a dramatic transformation in the second half of the century from being a major sending country in Europe to, by 1990s, a net recipient country of global migration flow. LA7, meanwhile, moved from a net receiving region to a net sender of the Global South.
Political and Economic Stability
For diverse reasons, Latin American dictators could not maintain the kind of stability that the Franco regime was able to impose. During the years leading up to and during the Spanish Civil War, the country was highly politically fragmented. During the period 1925–1939, Latin America was significantly less fragmented than Spain, with a fragmentation score averaging 44.4/100 for the period.Footnote 32 In other words, through the Second Spanish Republic and until the moment Franco assumed power, Spain was substantially more politically fractionalized than LA7.
A turning point occurred in 1939, and continued until 1976, the year after Franco’s death. During this period Spain witnessed preternatural political stability resulting from Franco’s authoritarian rule. Latin America, meanwhile, oscillated between the low 30s and high 60s on the political fractionalization index (although Brazil remained relatively stable and other countries, specifically Mexico, Peru, and Venezuela, are missing data) until the 1980s.
Until 1972–1973, Spain and LA3 (Argentina, Brazil, Mexico) had a current budget deficit per capita that hovered around zero (in current US dollars).Footnote 33 But then Spain began a period of intense deficit spending, causing the annual per capita deficit to grow prodigiously. By 1980, Spain’s budget deficit per capita had risen to $203.80. By 1990, it was $689.04, and almost $1,500 by 1995. In comparison, LA3’s budget deficit per capita averaged $78.05 in 1980, $96.52 in 1990, and $97.94 in 1995. In other words, Spain’s annual budget deficit per capita was fifteen times higher than the average for the three largest Latin American economies.
It’s clear that the Spanish economy enjoyed a price stability throughout the second half of the twentieth century that was rare in Latin America.Footnote 34 Price levels were relatively stable for both Spain and the three major Latin American economies in the years prior to 1950. Runaway inflation began affecting South America soon after, but Spain and Mexico continued to have nearly identical rates of inflation until 1980 – when the value of a basket of consumer goods in Mexico was 990 percent of its 1953 price level, and in Spain 1,151 percent. Then began the Mexican peso crisis. By 1985, the Mexican consumer price index (CPI) shot up to 10,609. Just five years after that, it mushroomed to 149,200 and kept growing, peaking at 230,834 in 1993! The Mexican experience pales still in comparison to the runaway inflation of Brazil and Argentina. As early as 1970, Brazil’s CPI hit 14,895 percent of 1953 values, while Argentina’s reached 3,774 percent. By 1980, the index values were 275,934 and 986,282 for Brazil and Argentina respectively; by 1985 they were 19.8 million for Brazil and 1.33 billion for Argentina. Only five years later, in 1990, Brazil’s price level was 422.8 billion percent of its 1953 value, and Argentina’s 18.45 trillion percent. In Spain, by comparison, the CPI relative to 1953 was just 2,033.7 in 1985, 2,770.9 in 1990, and 3,249 in 1993.
The differences in financial stability are also reflected in the premiums the countries had to pay for their debt. Between 1978 and 2000 (the years for which data is available), the average Mexican Treasury bill rate was 35.15 percent.Footnote 35 For Spain, it was 11.17 percent. This means that relative to Latin America, Spain had access to far cheaper external debt servicing, an important consideration in light of Spain’s high deficit spending noted above. The rate on Mexican T-Bills also fluctuated dramatically during this period, with a standard deviation of 23.25 compared with a standard deviation of 4.40 for Spain. After joining the European Union, Spain’s T-Bill rates dropped to single digits: 8.1 percent in 1994, 9.79 percent in 1995, and a low of 3.01 percent in 1999. Mexican rates also came down from their crisis-level highs, but continued to fluctuate: 14.09 percent in 1994, 48.4 percent in 1995 (a year of high global interest rates), and 21.4 percent in 1999.
Spain also had very low lending interest rates relative to Latin America as a whole, although this too fluctuated over time. On average, between 1977 and 2002 (and including the tumultuous 1980s), mean annual lending interest rates in LA7 were 112.53 percent – ten times higher than Spain’s average of 11.77 percent. The median Latin American rate, discounting outliers, was 42.91 percent – significantly lower, but still four times as high as Spain’s.
What about exchange rates? Compared with LA7, Spain did not always have the best exchange rates against the dollar, but even the most cursory historical analysis shows how much more stable Spanish currency was relative to Latin America. Between 1960 and 1980, Spain enjoyed remarkable currency stability averaging 64.58 against the dollar, with a range of just 19.10 and a single-digit standard deviation of 6.04. Among LA7 countries, only oil-rich Venezuela had a currency stability that came anywhere close to Spain’s. Mexico enjoyed the next most stable currency, although it too faced a sudden spike during the 1982 Mexican crisis, and in general did not come close to Spanish numbers. The Latin American economies also suffered sweeping currency swings in very short time frames, particularly during the turbulent 1980s.
Summary
The comparison between the largest Latin American economies and Spain suggests several possible observations. First, it confirms that the developmental state did not fail in Latin America; on the contrary, it produced impressive results well into the 1970s. This is especially true of measures of human development. The region did this while dealing with massive demographic change and political and economic instability. But Spain’s progress was certainly astounding, and it represented a profound social transformation. Why the difference in outcomes? We suspect two critical factors, which will be further analyzed in the book: greater institutional stability through the consolidation of professional state bureaucracies, and much easier access to foreign capital, permitting much greater deficit spending. To this we may add the geopolitical context which encouraged both Western Europe and the United States to provide as many opportunities and support as possible to Spain, in contrast with often counterproductive policies followed in Latin America.
All in all, the questions and puzzles presented in this introduction will be analyzed and discussed throughout the chapters of the book, and they will be re-examined in the book’s overall conclusion. For this purpose, we intent to follow a general plan of research, based on the four dimensions of state strength already defined above, which we will also consider in the next section, together with short summaries of each chapter.
Parts and Chapters of the Volume
The present volume begins, after this introductory chapter by the editors and Vivekananda Nemana, with a first part on Visions and Politics of Development, including Chapters 2, 3, and 4. Those three chapters of the first part do not focus –yet – on one or more of the dimensions of state strength in any particular national case. The developmental era was characterized by the circumstance, especially in Latin America, that relatively detailed strategies and “visions” for development were conceived and formulated by international organizations such as the UN commission CEPAL,Footnote 36 the World Bank, the Organization of American States, and others. The international organizations, foremost among them CEPAL, generated considerable resources, such as public policy programs, trained experts and their networks, and economic data, which national states could employ to support their own economic and infrastructural capacities. Moreover, economic policies and development experts gained national and international legitimacy – symbolic strength – as a result of being publicly endorsed by international organizations such as CEPAL, the Tennessee Valley Authority (TVA), the Alliance for Progress, and others. But the work and resources of international organizations did not form, at first, part of the capacities of any particular national state, until the state authorities decided to employ them – which they often did.
Chapter 2 by Joseph Love discusses the origins, intellectual history, and political influence of CEPAL on Latin American countries, and also on Spain and Portugal, between 1950 and 1990. The chapter focuses on “Structuralism” as a school of thought that was central for CEPAL’s economic analysis and public policy proposals. The chapter considers other contributions of CEPAL that are very rarely discussed in the literature, such as blueprints for institutional design of developmental agencies, and programs for training of public policy experts and high-ranking civil servants. Love also describes briefly the attempt by Celso Furtado, who was trained at CEPAL, to implement some of the ideas of the commission, acting as a leading public policy expert and politician during the early 1960s in Brazil. Finally, the chapter discusses how CEPAL reacted to the challenge represented by the ascent of the neoliberal consensus in the 1990s.
Margarita Fajardo in Chapter 3 takes further the analysis of the difficult – and sometimes dangerous – political implementation of CEPAL’s vision and strategies for development “on the ground,” considering again in particular the case of Brazil. The chapter focuses on sociologists and economists, and their academic and political engagement to promote development as a long-term national project. The chapter shows that intellectuals, academics, and public policy experts were initially enthusiastic and optimistic as regards their potential for creating and leading state institutions that were going to advance social and economic development. However, the initial optimism turned relatively soon to political disorientation, and theoretical bafflement. Since the middle of the 1960s, intellectuals and experts realized – sometimes already in exile – that it was necessary to think again about the sociological foundations of state power, and about how to reconcile capitalism with democracy in Latin America and Spain.
Chapter 4 by Robert Karl discusses another key source for developmental visions and strategies in Latin America, particularly in the case of Colombia: the TVA. Created in 1933, TVA was regarded not only as a blueprint for institutional design as a public agency, but also as a model for the public policy strategies it applied – with great success – to promote the economic and social development of a vast geographical area in the United States. After World War II, the TVA model was studied and applied by several national governments in Latin America, it was promoted by the World Bank, and it became influential on the development vision championed by the Alliance for Progress. The author describes the fact that Colombia was an early “showcase” for US American development agencies and experts. However, due to a series of difficulties and setbacks, the country found itself relatively soon considered as a “failure,” even by many – or most – Colombian sociologists and economists. Development programs seemed never to fulfill the high expectations that surrounded them at the beginning.
José Carlos Orihuela, in Chapter 5, begins the part of the book that focuses on the study of infrastructural and territorial power. The volume has been similarly organized for each part: the initial chapter compares several national cases, and the rest of the chapters present each one national case study. Orihuela’s chapter compares three developmental agencies in Chile, Colombia, and Peru. The blueprint for the three national agencies was the TVA, the flagship developmental agency of the New Deal, which is also discussed in the previous chapter – as well as in several others, and again particularly in the conclusions of the volume. This same institutional blueprint was “translated” in very different ways by the political and institutional actors of the three countries. The chapter examines the reasons or factors that explain the bureaucratic strength of each of the agencies thus created, considering not only their autonomy and professionalism, but also other significant factors such as political legitimacy and embeddedness.
Chapter 6 by Eduardo Dargent examines a key developmental agency in Peru, the Instituto Nacional de Planificación (INP). The INP was created in 1962 following blueprints suggested by the Organization of American States (OAS) and CEPAL. The author evaluates the institutional project of INP as failed, but the case is very revealing nonetheless, and the author examines on its basis diverse theoretical models that have tried to explain the emergence and endurance of state capacity in Latin America. The initial phase of INP was already very problematic, because evidence points out to the fact that the organization was originally conceived as a “Potemkin institution.” This interesting characterization of certain institutional projects in developing countries was first formulated in Russia, of course. The chapter pays special attention to informal political practices as the foundation for bureaucratic autonomy, as suggested by Carpenter’s theory of agency reputation, among other relatively recent contributions.
Luciana de Souza Leão, in Chapter 7, analyzes the many efforts and projects that aimed at strengthening and expanding both the infrastructural power and the territorial reach of the federal state in Brazil from 1930 to 1985. The chapter begins with the creation of a career civil service during the era of President Vargas, in the late 1930s, and goes on to consider other projects of institutional modernization in the following decades, including the creation of independent developmental agencies. The author discusses the fact that bureaucratic autonomy, in Brazil, was never understood to mean real political independence for public agencies, but rather their direct subordination to the president. This institutional design facilitated frequent political interference on public policy decisions, and their erratic character as a result. The chapter examines also the diverse projects of bureaucratic modernization that involved expanding the territorial reach of the Brazilian state with the creation of regional development agencies.
Chapter 8 by Agustin E. Ferraro and Juan José Rastrollo discusses the correlation between professional bureaucracies and economic development, first postulated by Weber, and the development of the Spanish industry as an instance of this correlation during the developmental era. First of all, the authors consider the reasons that explain the curious lack of attention to the Spanish case in the literature on economic development, even among studies of development in Latin America. Secondly, the study of a specific economic area, the automobile industry, shows that state-led industrial promotion was very effective during the period 1950–1990 in Spain, and that the effectiveness of industrial policy was substantially increased after the modernization and professionalization of the central state bureaucracy in the early 1960s. The chapter argues, finally, that the legal and institutional blueprints for civil service reform in Spain, implemented during the period of state modernization and (controlled) transition to democracy between 1960 and 1978, had been originally established during the era of profound social and political modernization of the country from 1914 to 1936.
In Chapter 9, Jordi Catalan and Tomàs Fernández-de-Sevilla begin the part of the book focused on the state’s economic power. The chapters in this part consider also bureaucratic structures and the territorial reach of states, certainly, but they pay particular attention to the economic public policy framework, including labor, welfare, and fiscal policies, and their impact on diverse areas and variables, such as industry, trade, growth, public debt, and others. As mentioned above, the first chapter in each part adopts a comparative perspective, and discusses several national cases. In Chapter 9, the authors examine and compare the sometimes fluctuating developmental strategies, and the macroeconomic evolution for three countries: Argentina, Brazil, and Spain. The chapter examines the often conflictive political context for economic decisions, the long-term consistency – or lack thereof – in the implementation of developmental programs, and their impact on the main economic variables during the period under study.
Chapter 10 by Alan Knight examines the general political context, the economic policies, and the state building programs during five periods of Mexican history, from before the Revolution to the end of the developmental era. The first period, the Porfiriato, witnessed a deterioration in welfare, fall in real wages, and rising inequality. The revolutionary generation of the 1920s and 1930s began building a stronger state, with enhanced powers in areas such as commercial and labor regulation. However, the most successful growth model was reached during the period from 1945 to 1972, under the semi-authoritarian regime of the Revolutionary Institutional Party (PRI). This period included effective long-term, state-led industrialization policies. During the late 1970s, the PRI regime tended to overreach in terms of statism, nationalism, and economic populism, leading to the debacle of the early 1980s.
Yovanna Pineda, in Chapter 11, analyzes import-substitution policies in Argentina, from the early 1940s until the end of the developmental era in the country. Across different periods and political regimes, the chapter focuses on a particular area of production, the farm machinery industry, in order to evaluate the economic impact of the – frequently – shifting public policy frameworks. Based on deep interviews with key informants, associated directly or indirectly to a traditional harvester manufacturing firm in the Province of Santa Fe, the chapter also explores the sociocultural impact of developmental initiatives. The interviews provide crucial insights into the experiences and beliefs of firm owners, engineers, farmers, and their family members, including their perceptions of public officialdom, and of distant government decisions. Finally, the author considers a relatively recent attempt to reestablish industrial developmental programs in Argentina, and the sudden protagonism that the agriculture machinery industry came to assume in one of the episodes of this failed political project.
Chapter 12 by Patricio Silva presents a study of the Chilean developmental state that begins by considering the country’s nitrate-based economy during the late nineteenth century, and its political correlation, the Aristocratic Republic. The decline of the aristocratic regime, and the comparatively early establishment of developmental institutions and strategies during the second half of the 1920s, resulted from three main factors: the decadence of the nitrate industry after World War I, the increasing discontent with the aloof ruling elite, and the eruption of the middle class on the political scene. Already in the second half of the 1920s, new specialized state agencies were created to manage policies of wide credit support for central areas of national production, including mining, agriculture, and industry. Nevertheless, the Chilean development state consolidated as a national project only in the late 1930s, with the establishment of Corporación de Fomento de la Producción (CORFO). The Chilean developmental state became one of the most successful in Latin America but, as the author shows, the stability of economic policies depended on political compromises between the main political coalitions of the left and right. The increasing political polarization since the early 1960s, and the loss of the willingness to reach stable political compromises among political actors, had catastrophic consequences not only for the developmental project, but also for Chilean democracy.
Matthias vom Hau in Chapter 13 begins the part of the book focused on symbolic state power, including discussions of national and civic identities, their relationship to the developmental state, and their transformations during the developmental era. As the first chapter in this part, the study presents a comparative analysis of popular nationalism and the developmental state in three national cases: Mexico, Argentina, and Peru. Nevertheless, as the author indicates, the chapter’s main theoretical claim can be applied to other national cases, as shown by brief references to Brazil and Bolivia. Vom Hau argues that popular nationalism was neither a mere legitimation tool of the developmental state, nor completely unrelated to it. Instead, he suggests that the concept of “selective affinity” works better to describe the relationship. Originally employed by the writer and statesman Johann Wolfgang von Goethe to analyze human relationships, the concept of selective affinities was first applied in the social sciences by Max Weber, who famously postulated a selective affinity between Protestantism and capitalism. The chapter shows that, far from being a mere legitimation tool of the developmental state, Latin American nationalism had sometimes a conflictive relationship with it.
Chapter 14 by Marshall C. Eakin examines the powerful Brazilian narrative of cultural and racial mestiçagem (miscegenation) that was promoted by diverse social and political actors since the 1930s in the country. The author argues that the developmental state in Brazil, after 1930, played a central (but not determinant) role in the construction and success of this narrative, that the narrative played a critical role in the gradual emergence of Brazil as a robust nation state by the 1970s, and that the power of the narrative spanned political parties, social movements, classes, and regimes. The narrative of mestiçagem was at the core of cultural nationalism in Brazil, and also helped foster a dynamic civic nationalism by the 1980s and 1990s. Finally, the author claims that the appeal of the narrative continues, but that its potency has eroded over the past two decades. At the end of the chapter, the Brazilian narrative of mestiçagem is briefly compared with dominant narratives of race and nation in other Latin American countries.
Brodwyn Fischer in Chapter 15 discusses urban informality, and the paradoxes of its relationship to economic development and political citizenship during the twentieth century in Brazil. Often described as development’s negation, the author claims that urban residential informality has played a much more complex role than such simplistic analysis would suggest. On the one hand, urban informality has been a vital escape hatch from a long series of developmentalist dilemmas: in economic, legal, and political terms, it has facilitated the adoption of ambitious policies that far outstrip available resources and capacities. On the other hand, urban informality has served to reinforce deep economic and civic inequalities, perpetuating historical forms of marginalization, and facilitating the creation of layered, contradictory, and often destabilizing economic and citizenship regimes. Urban informality has shown the capacity of the developmental state to redefine citizenship, but at the same time it has put in evidence its limitations.
Chapter 16 by the editors summarizes the results of the volume, and includes a concluding discussion, as well as attempted answers, for two related issues or questions raised in this introduction, and further examined throughout the book: first of all, the reasons for the relative lack of success of development programs under technocratic-authoritarian regimes during the 1960s and 1970s in Brazil, Argentina, and other Latin American countries, compared with their remarkable achievements in Spain; and secondly, the institutional design, modes of operation, and political practices adopted by developmental agencies under democratic regimes, which we can, in retrospect, evaluate as contributing factors to their effectiveness in promoting economic growth and social progress during the developmental era.
